MSF Feature Request: Add S&H Modulators to the per-voice modulators

Official support for: meldaproduction.com
RELATED
PRODUCTS

Post

The goal is the following simple scenario:
1. Setup a free running, slow (0.25hz) ramp up LFO (Use Modulator in Utility section for that).
2. Route it thru a S&H module which takes samples by key press and holds until release
3. Use the output of the S&H as modulation for filter cutoff
4. setup the filter resonance to near self oscillation.

Play this with a 1/16 sequence.
What you get is a step by step increasing filter cutoff up until the LFO ramp cycle starts again.
Reference: Kraftwerk doing his with the Odysse

Solution:
1.) Add a number of S&H modulators along with the existing Env, Lfo, Random.
2.) Options for "When to take sample":
- Keypress
- Other Modulator such as
-- ENV exceeding a certain level
-- LFO hitting max value, think of a square wave LFO...
3.) Options for Inputs from which to take samples to be held
- LFO
- ENV
- Modulators
That could be done like with the "Set" module where a "Value" could be modulated by a "global Modulator" or by a per voice one.

BTW: With this approach the current "Random" modulators could easily be emulated by a white noise LFO and Sample-Mode "keypress"

Link to Reference (kraftwerk home computer, at around 1:43) : https://youtu.be/bu55q_3YtOY?t=103
Last edited by ] Peter:H [ on Sun Jun 09, 2019 9:19 am, edited 1 time in total.

Post

1. Setup a free running, slow (0.25hz) ramp up LFO.
The Per Voice LFOs can't be set to free running. Scroll to 4th post on this page and read on:

viewtopic.php?f=1&t=524931&start=930

Post

@David - thanks I know, I was talking about the Modulator LFOs in the Utilities section.
These can be run free. That's Melda standard.

It's interessting to have such a global LFO which somehow "gives global direction" to a modulation.
Then by sampling the very current value on a per voice basis each individual voice get's an "individual touch" but nevertheless takes part in "the global direction" of the modulation.

Global Utilities LFO --> Sample&Hold on per voice basis / Key-Press --> Mod Filter Cutoff

Compare this to for instance Bitwig 3 "The Grid": I can set this sort of modulation up in 2 minutes in "The Grid". I think it should be a.) possible in MSF and b.) it should feel "natural" to people who know synths and have some experience in modular environments.

Anyway - long story short - For me this is kind of a litmus test for this modular environment: I wanted to reproduce a kraftwerk patch of the late 1970ties which they did with a analog Odyssey synth. I spend now quite a few hours with this simple thing in MSF and though I use Melda tools for quite some time (Owner of MPS...) I'm just to stupid to get what I want.

Therefore my conclusion is: MSF would profit from S&H Modulators.
You do not have the required permissions to view the files attached to this post.

Post

MSF would profit from S&H Modulators.
Ditto! (Thanks for the clarifications:)

Post

A S&H module and a quantizer could be so useful: an lfo or any source send to a sample and hold module trigs by another source, like an lfo....the output is quantize by a musical scale........like in modular or in Hive2! Result: polyphonic controlled generative madness.
Best
YY

Post

Code: Select all

$eNqlmdty2zYQhu-1FBj6NokALI4zlDpxIiWe2rFquWl7SUuwzQlFqiTlsfv0HQAkpVg2zJY3Gi3wfwssTiQX8S+Pmww9mLJKi3wSkQ84QiZfFes0v5tEu-r2vYp+mY7ii2Wxy9fzZFUX5VNl6jrN7yp0MolIhE6+ZMVNklXXyY2zP+3K0uT1ojSVqc8+T6Jlstlm5mO+-lpk69nj1pTpxuR1hGbrtC7KZfqPmUSEUfFOKDgsnSXV0yQimMp3EliEvv72+7ayzlzvcITOr5bbLK1r4+TEavlRMZ1ElOkIXSSP5-PL06xY-fBNAo3Q9zZ0xSlTOJrGy93NItvdpXkT7TT25vLnqF-0tjSZWdVm7YlJhN+JqOWr9g+a77LsW7Ixk+iyWqVZltRFiZzL6sdZ7gJ7Xh2hP135X5OIRa-06OT3PK2KfJYnN5n5VGRZsq3M+tXOTuOLy88nS7NJ6yI3FfpstvW9VY+n8fI+2Rq0vJ4tFlez5ezatf2QZDs7UWAV45+70BW8EOQ8zWrjAjy0DyISr0Y0n0TvyQdOVYROrp+2ZhKdLxBloYDm6FOxy2vHNyHhD0za6it8UFeVKzyJzueXyEU8dmwornE3jceaozUzOlhGf74S3UsxvLCC8H4FvTjuhw2N4ln+QKbxdWk2RVa8NI3WhxM5Le2jpV4LfbTgtayPlnkt76PlXiv6aIXXyj5a6bWqj1ZNR7E-62b5w5vAXjmKz+eXpF17QCLk96jfm+cmv7OrFCJ0ldTGLlcMlGEAJoVmpFnXtg5dl0le3RblJqnTdqc7L7Y3h1Vo9u172x3drW-rw3s7y9M6TbLFfVKZ4w1DolBLP22jw0j8lvpW1AaVSb4uNui4Y3gaL4o0r9E8S+6qU3vuu+E6LkSPzVBQrrDg-UToqT8kGCeCUD0I6uwQpIWUjGg2COrsAESo9mEMgfZ2CBJeBoOgzg5BGjhToOQgqLMDkJ9-xukQaG+HIK8hYhDU2SGomRYyCOrsAARtEEOgvR2C-KyQIUxrhhAl-f4ZwnR2AGLNQtNDoL0dgnizewZBnR2ClI9CDYI6OwC1jukQaG+HoMa1HAR1dghqNikZBHV2AGoioGIItLdD0M9Pzv-F9Hhu+gcecD6E6ewAJNuH6xBob4egZpuyQVBnh6BWNQjq7ADULjMYAu3tENQep4Ogzg5BzcTQQVBnByDdvqwOgfZ2CHr24Px-UJ9Hp2430CCos1+FyH+V+vzA0WfM+Ar3++o6-NaaxsvabNFZbTbYPeGYJgQIxlLIyBUTdwAJAMZBC25PCleMJ9F7-AErJoFLqhS2z2NXQ1wNAAHMCAYCoqmg7t2NKSGxwsS+obpicMuMSS6wFAC8bYE5PxxTLYBRkLSt4O4NWjGiADNBaetI+DOZCUFACPul4oqlG10tORECKFZNsXLuNQFQgmDBNGkqtDs0NKdSE8Ypa8qp-+pRIAUI4KTtJ-UjQQVRSnENzE6+q-ADQRjByjqCZkSp8yQ5w1gTJbv+U-A9YlJKyYWmbQTUj4SQAgMVgoJuCe4qmAaMGSHA2qGmwr+uKMa0EFTT1pP0ACMgMdZaqmbSqPJLkEstJaXQ6t1YCEUFlhSAt36gOe0ABFdE4iYycGvIdl0RyYWSzVizZkVQLbmmDJNmhHz3sZAaBKGUEdn4d92XimpGgTMhm6F2M0m4JAw4uM82W+onUgLjmmhMADejo32rioAGIhhoESG72vdJBWtZ5T4TYveITY74HAntkiE2E-Yso+IU8KaCvangbyrEmwr5pkKFFVcuQUJCJ4h7FI5bYYvQvgjtEOiLQIewvojN4f1qnqpVkpke0eyle4z2x2xM301WrNK6d5PP9M8c0P-o4LAHz6WjvSt7+VHtShOSpPXq-tTk64Dmolj-cW9MFpBcp5ub0kgWkJyWJqnvMQ1IZo9b2+O0yAkJyGxu7-L29ljxJUvXxmdCj+rO55fV1P5iNC-N3zuTr558ArxYG5svLDdJZpPcSZlsjL0qsHcHV0l+Z+xtB+rK-fWBM+1tgmvuMP-43V9Q4ObfabL6sdu6EzJqrzSOJrdzPo2X6V2eZEd7NV4+5auDzdwcWfYNwEU2ij+u6vTB+Psu1F68XJfG98VeZ-nE6aekNndF+WSPwLP8thhP4+au7JjysdrGPi68Z9vuYUu24XLb5Ixtt0bxhQ2kotP4j3Rd319M42WxK1cGLUpz6y5dmnsz66qVjFtoFI9fuuWrpqN-AX3hpRA=
Here's a minimal preset that implements a pseudo-random S&H LFO in the generator. The same thing could pretty much be done with a global Mod unit as well.

LFO1 modulates filter cutoff with a pseudo-SandH modulation.

A random 31 step-pattern is used. The LFO start phase is modulated by a random number i=n/31 for n randomly equal to 0, 1, ... 20, 31. In other words, the starting step in the step pattern is random. The trick is the transform grid taking the random number and translating it to a n/31 value for the start phase. Thus a random step in the step-pattern should be choosen. While not truly random, this is a far-from-obvious fake-random pattern.

As long as note-on is on a sync beat (i.e., LFO length = 1/8 and the envelope is triggered on eight-notes or multiple, eveything nicely syncs up. But if note-on is between LFO-length, you lose the on-beat coordination, unfortunately. I couldn't figure out how to nail that down.

The number of steps is 31 because it's the largest number of steps that won't repeat on the downbeat of a measure any time soon. If I could have specified 63 or 127 steps, I would have.

Post

I like it, thankyou!

Post

dmbaer wrote: Tue Jun 11, 2019 11:56 pm Here's a minimal preset that implements a pseudo-random S&H LFO in the generator. The same thing could pretty much be done with a global Mod unit as well.

LFO1 modulates filter cutoff with a pseudo-SandH modulation.
[...]

The number of steps is 31 because it's the largest number of steps that won't repeat on the downbeat of a measure any time soon. If I could have specified 63 or 127 steps, I would have.
Thanks dmbaer, nice creative solution. Would you nevertheless support adding S&H Modulators to MSF or do you think your solution meets the cirtieria "feels natural to modular synthesists"?

Post

] Peter:H [ wrote: Fri Jun 14, 2019 4:22 pm Thanks dmbaer, nice creative solution. Would you nevertheless support adding S&H Modulators to MSF or do you think your solution meets the cirtieria "feels natural to modular synthesists"?
I would love to see an S&H (i.e., random) option for the per-voice LFOs, make no mistake. I'd also love to see a random drift LFO capability as well. And I'd like those to be selectable right along with other standard LFO waves from a bank MP.

I think the work-around solution I presented would fool most listeners into thinking they're hearing an actual random LFO modulation, apart from the sync issue, but it's way more work than I'd prefer to do.

As to a full-blown S&H capability based on some kind of user selectable side-chain(ish) signal ... well, maybe that's something modular enthusiasts would know how to effectively use, but I'm not sure I would ever feel the need.

Post

I am missing something here. Please explain it to me.
So, my understanding of Sample and Hold (S&H)
Is that it is just a randomizer.
I struggle to see how that is different from the randomizers we already have in MSF.
Could you please try to explain how S&H is different.
Melda Production & United Plugins
Surface Studio = i7, 32gb, SSD.
Windows 11. Bitwig, Reaper, Live. MTotal.
Audiofuse, Adam Audio monitors + sub, iLoud MTM.
Polybrute, Summit, Pro 3, Tempest, Syntakt, AH2.
Ableton Push 2, Roli Seaboard Block.

Post

jmg8 wrote: Sat Jun 15, 2019 9:10 am I am missing something here. Please explain it to me.
So, my understanding of Sample and Hold (S&H)
Is that it is just a randomizer.
I struggle to see how that is different from the randomizers we already have in MSF.
Could you please try to explain how S&H is different.
S&H is not a randomizer. A S&H module is usually feed by an input source.
Sources can be arbitrary signals like an LFO.
From the Source the S&H takes samples with a given time raster, like 1/8.

A set up where S&H is very different to random is
Source: Free Running LFO, Ramp up waveform, LFO cycle takes close to 2 full bars
S&H Raster: 1/8

That means while the free running LFO climbs up, the S&H module takes a sample each 1/8 and holds the value. This means you get a stairs like graph if you would draw it, i.e. output values of the S&H would increase one by one, up until the LFO cycle starts again. Compare this to random - it would give you value sometimes higher sometimes lower than the previous one

Difference to using the LFO itself: Say you want to s&h with raster 1/4.
And S&H should control filter cutoff
Using LFO: While holding a note for 2/4 you get a continuous filter modulation
Using S&H sourced by same LFO: You get 2 distinct values which are s&h

Hope that makes it a bit clearer...

BR
] Peter:H [

Post

So if a sample is taken at every 1/8 and the key is held down for a bar, is the value changing every 1/8 or is it held for the duration of the key press?
Melda Production & United Plugins
Surface Studio = i7, 32gb, SSD.
Windows 11. Bitwig, Reaper, Live. MTotal.
Audiofuse, Adam Audio monitors + sub, iLoud MTM.
Polybrute, Summit, Pro 3, Tempest, Syntakt, AH2.
Ableton Push 2, Roli Seaboard Block.

Post

jmg8 wrote: Sat Jun 15, 2019 4:27 pm So if a sample is taken at every 1/8 and the key is held down for a bar, is the value changing every 1/8 or is it held for the duration of the key press?
Depends on the "When to take sample" setting. The example in my previous post used the "Sync to song" setting. And in this example, it's the first option "the value of the source is sampled every 1/8".
But a sample could be taken on every key press (gate signal in modular terminology) as well, that was your 2nd example.
Or it is taken on ARP note on (same like gate signal).
Or when another LFO reaches it's signal peak.
There's numerous options for "when to take sample" and "from which source take the sample"

BR ] Peter:H [

Post

I'm not asking about when the sample is taken, I wanted to know about already held notes. Do they just keep the current sample for as long as it is held?
Melda Production & United Plugins
Surface Studio = i7, 32gb, SSD.
Windows 11. Bitwig, Reaper, Live. MTotal.
Audiofuse, Adam Audio monitors + sub, iLoud MTM.
Polybrute, Summit, Pro 3, Tempest, Syntakt, AH2.
Ableton Push 2, Roli Seaboard Block.

Post

jmg8 wrote: Sat Jun 15, 2019 5:31 pm I'm not asking about when the sample is taken, I wanted to know about already held notes. Do they just keep the current sample for as long as it is held?
The S&H holds a value up until the next value is sampled...I the sample-trigger determins when the sample is taken and for how long it's been held, i.e. until the next sample trigger is reached. There's no two different things, like the sample rate and the hold rate...
Probably you own Bitwig and do beta testing for Bitwig 3 the Grid? The S&H module has two inputs, a.) the source from which to take samples and b.) a gate input which can be feed by arbitrary signals.
It's depicted here: viewtopic.php?f=138&t=526583

Post Reply

Return to “MeldaProduction”